Hugging Face's logo

RichardErkhov
/
RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf

GGUF
Inference Endpoints
conversational
Model card Files Community
RichardErkhov commited on
Commit
6c83f6a
•
1 Parent(s): 1295230

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +109 -0
README.md ADDED
@@ -0,0 +1,109 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1 - GGUF
11
+ - Model creator: https://huggingface.co/RyanYr/
12
+ - Original model: https://huggingface.co/RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q2_K.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q2_K.gguf) | Q2_K | 2.97GB |
18
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_S.gguf) | Q3_K_S | 3.41GB |
19
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K.gguf) | Q3_K | 3.74GB |
20
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_M.gguf) | Q3_K_M | 3.74GB |
21
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q3_K_L.gguf) | Q3_K_L | 4.03GB |
22
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.IQ4_XS.gguf) | IQ4_XS | 4.17GB |
23
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_0.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_0.gguf) | Q4_0 | 4.34GB |
24
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.IQ4_NL.gguf) | IQ4_NL | 1.23GB |
25
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K_S.gguf) | Q4_K_S | 4.36GB |
26
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K.gguf) | Q4_K | 4.57GB |
27
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_K_M.gguf) | Q4_K_M | 4.06GB |
28
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_1.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q4_1.gguf) | Q4_1 | 4.77GB |
29
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_0.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_0.gguf) | Q5_0 | 5.21GB |
30
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K_S.gguf) | Q5_K_S | 5.21GB |
31
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K.gguf) | Q5_K | 5.33GB |
32
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_K_M.gguf) | Q5_K_M | 5.33GB |
33
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_1.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q5_1.gguf) | Q5_1 | 5.65GB |
34
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q6_K.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q6_K.gguf) | Q6_K | 6.14GB |
35
+ | [self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q8_0.gguf](https://huggingface.co/RichardErkhov/RyanYr_-_self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1-gguf/blob/main/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1.Q8_0.gguf) | Q8_0 | 7.94GB |
36
+
37
+
38
+
39
+
40
+ Original model description:
41
+ ---
42
+ base_model: mistralai/Ministral-8B-Instruct-2410
43
+ library_name: transformers
44
+ model_name: self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1
45
+ tags:
46
+ - generated_from_trainer
47
+ - trl
48
+ - dpo
49
+ licence: license
50
+ ---
51
+
52
+ # Model Card for self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1
53
+
54
+ This model is a fine-tuned version of [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410).
55
+ It has been trained using [TRL](https://github.com/huggingface/trl).
56
+
57
+ ## Quick start
58
+
59
+ ```python
60
+ from transformers import pipeline
61
+
62
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
63
+ generator = pipeline("text-generation", model="RyanYr/self-correct_Ministral-8B-Instruct-2410_metaMathQA_dpo_iter1", device="cuda")
64
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
65
+ print(output["generated_text"])
66
+ ```
67
+
68
+ ## Training procedure
69
+
70
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yyr/huggingface/runs/rdf560eo)
71
+
72
+ This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
73
+
74
+ ### Framework versions
75
+
76
+ - TRL: 0.12.0.dev0
77
+ - Transformers: 4.45.2
78
+ - Pytorch: 2.4.0
79
+ - Datasets: 3.0.2
80
+ - Tokenizers: 0.20.1
81
+
82
+ ## Citations
83
+
84
+ Cite DPO as:
85
+
86
+ ```bibtex
87
+ @inproceedings{rafailov2023direct,
88
+ title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
89
+ author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
90
+ year = 2023,
91
+ booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
92
+ url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
93
+ editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
94
+ }
95
+ ```
96
+
97
+ Cite TRL as:
98
+
99
+ ```bibtex
100
+ @misc{vonwerra2022trl,
101
+ title = {{TRL: Transformer Reinforcement Learning}},
102
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
103
+ year = 2020,
104
+ journal = {GitHub repository},
105
+ publisher = {GitHub},
106
+ howpublished = {\url{https://github.com/huggingface/trl}}
107
+ }
108
+ ```
109
+